Event-based Lateral Security Control for Network Autonomous Driving System Subject to Double-Quantized Scheme

Abstract

This paper investigates the event-based lateral security control problem for network autonomous driving system subject to the ideal path tracking performance, in which a novel double-quantized structure is reasonably integrated into a unified autonomous vehicle model. Firstly, a generalized tripled cyber-attack protocol, including deception attacks, DoS attacks and replay attacks is taken into  consideration in this paper to simulate the interference of the external environment. Secondly, in order to improve the data transmission efficiency and relieve the effects of cyber-attacks, asynchronous even-triggered scheme (AETS) and doubled-quantized-based control protocol are introduced simultaneously. Additionally, with the help of elegant linearization technique, which not only makes the augmented closed-loop systems global exponential stability can be achieved, but also the desired output feedback controller gains and triggering parameter matrices are co-designed to ensure the H\(\infty\) performance for the autonomous driving system. Finally, a practical example is given to verify the effectiveness of designed mentality.